To extend or replace: That is the question

September 2nd, 2014, Published in Articles: Energize


All power plant has an operational lifetime  and at some stage will reach the state where it no longer meets performance requirements. At this point the decision must be taken whether to replace or extend lifetime by rebuilding, overhaul or other means. This applies at the moment to existing aging utility power plant but will in future also apply to independent power producers who will be subject to a different set of constraints from the utility, and may not have the same options available .

The IRP 2010 has factored in the retirement of several aging coal fired stations but decisions on aging power plant can be affected by a constrained generation situation where demand exceeds capacity. The main focus at present is life extension of existing coal fired plant. The design life of large power stations is generally considered to be 40+ years with 60 years for nuclear and some claims of 100 years design life appearing. Medupi is claimed to have a design life of 50 years.  A long lifetime is necessary for large power plant to recover capital costs while keeping generation costs low.

Utilities generally have the following characteristics that may affect the decision:

  • Long term or permanent generation licences
  • Power plant with long design lives
  • Regulated prices
  • Large network of power stations at its disposal, established technology and extensive field experience.
  • Ability to accommodate long projects

A utility such as Eskom is required to keep prices low, but is not under any contractual restriction on costs, or subject to competition. The price determination is also based on a pass through of unavoidable costs, such as those incurred by compliance with legislation etc, which  may affect the decision to extend rather than replace

New entrants into the market, the IPPs, have a different set of parameters. Privately operated plant is far away from retirement age but will still have to face the decision at some stage in the not too distant future. With suggestions that the actual lifetime of plant in the field could be much shorter than the design lifetime, the choice may be facing them sooner than thought.  Situation is different for a utility operating under a price regulated regime than for a private operator under market or contracted price, and the following characteristics may apply:

  • Short term contracts – typically twenty years, with possible extension
  • Fixed prices based on a bidding system. This forces down prices which eventually means cheaper equipment with shorter design life
  • Smaller plant with multiple units on site
  • New technology with shorter design life and little field experience

New technologies such as solar PV and wind are being installed in large quantities and usually under tender conditions, which force down the price of plant, and which may affect the lifecycle. Plant owners need to take these facts into account, when looking at operation of plant beyond the contracted period of guaranteed payments. Other new technologies are also subject to limited time contracts under which guaranteed payment is made. Extension of the lifetime of the plant may depend on extension of contracts and contract prices. Contract extension may involve new conditions, such as new pricing and new off-take agreements, and the operators may find themselves having to participate on a short term bidding or spot market basis, rather than guaranteed off-take.

Decision triggers for the IPP may be:

  • Inability to supply contracted level of power or amount of energy
  • High operational costs, reducing margins at fixed prices
  • High downtime and maintenance costs

Operational costs are particularly sensitive to maintenance and repair costs in the case of power plant with zero fuel costs, such as wind and PV, and reduction of output due to downtime and non-availability affects revenue without reducing running costs. Life extension in a fixed term contract depends on how far the contract has to run and the possibility of extension. In some cases replacement may be the only option available.

There is in addition a third class of generation which will be faced with the decision, namely the own generator, who generates power for own consumption, with possible feeding of excess power to the grid. This class is faced with a different set of criteria – characterised by small plant consisting of either rooftop solar, gas engines or cogeneration plant. Plant usually consists of a few hundred panels and one or two machines. Decision triggers may be:

  • Reduced production leading to increased use of grid power, or decrease in net metering benefit
  • Increased operations cost- affecting the financial viability of own-generation
  • Reduced reliability and increased downtime, affecting security of supply, which is one of the main reasons for using own generation

The decision for the own generator is not subject to the same tight restrictions as the IPP , as there is no limited period contract involved

End of life criteria

End of life has been defined as “The useful life of a power plant is that after which repairs become so frequent and extensive, that it is found economical to replace the plant with a new one” [1]. This is considered a rather simplistic approach as many other factors such as constrained generation play a part in the decision.

End of life or end of useful life may be described as the point at which a power plant no longer meets required performance criteria with normal maintenance operations. Criteria may include:

  • Efficiency – the amount of fuel consumed for the required output
  • Power output capability – the capability to deliver electricity at the required power level
  • Energy output capability – the capability to deliver required amount of energy in required time period
  • Availability and downtime-out of service periods due to maintenance and failures
  • Operations and maintenance costs – frequent and extensive maintenance push up costs and reduce availability
  • Environmental requirements – compliance with emissions, pollution and other environmental requirements

In addition to existing performance requirements, future requirements, such as more stringent environmental requirements can add to the criteria, when life extension is considered. Also new technology may become available with improved performance

Lifetime of various technologies

Table 1 shows the projected lifetimes of various technologies used in the renewable energy and IPP market.

Extend or replace 1

Table 1: Projected lifetimes of various technologies used in the renewable energy and IPP market.


(a)    Actual field lifetime as reported in various sources depends on a number of environmental parameters such as ambient temperature, wind speed, humidity and others. Also there is a range of lifetimes of different modules on the same site, with wide variations being reported. Manufacturers tend to report on the best performance  and the average and ignore the worst.

(b)   Thin film modules seem to be subject to corrosion of the layers depending on environmental conditions.

(c)    Continuous operation – time to major overhaul

(d)   Rotor blades, gearbox and alternator replacement

(e)    Equivalent operating hours – determined by operational cycles

Replacement decisions

Replacement or extend decisions are usually taken on the basis of economical analysis of the cost of keeping plant running vs the cost of replacement. This assumes that it is physically possible to replace plant without impacting on the security of supply. A constrained generation situation may well favour extension in spite of basic economics showing otherwise, and the decision to extend life rather than replace is subject to a different set of constraints.

In the new segment of the market opening to independent operators and based on tender/price bidding system, with limited contracts and equipment with relatively short design life, the decision criteria used to generate an analysis will be significantly different.

System characteristics affecting decisions

One of the big differences with IPPs that a plant may consist of many small generating units. This runs into thousand of units in the case of solar PV, hundreds of units in the case of wind farms, tens of units in biogas plants using IC gas engines and possibly tens for units for fluidised bed coal or biomass plants. This affects the ability to detect degradation of performance and decide which units to replace.

In the case of solar PV at best individual strings are monitored. System performance is not necessarily an indication of module performance as system degrades slower than modules. Components do not all age or deteriorate at the same rate, and the life achieved under operational conditions will depend a lot on how the plant has been operated.

Passive systems

Renewable energy has added another dimension to the question, that of passive systems, where the rate of aging will depend more on the effects of the operating environment, such as temperature, than on operational  regimes or usage.

Solar power systems

The output of solar panels is known to deteriorate with time, and the decision here is a rather unique one in that the option of repair does not exist, the only option being replacement. The deciding factor is deterioration of performance of modules to a predetermined level, generally 80%. A difference exists between module degradation and system degradation with system degradation being lower than module, showing that modules do not all degrade at the same rate, and the system adjusts its performance to cater for changes in individual components [2].  This affects the ability of the operator to determine the status of the modules.

Two options must be considered:

  • Total replacement of all modules in one shot. Replacing with modules of higher efficiency provides the opportunity for increased production and hence increased revenue, vs a smaller number of replacement modules.
  • Extension of the existing DC side with sufficient new modules to compensate for deterioration of older modules, and progressive replacement of older modules. This depends on the availability of space for extra modules.
Extend or replace 2

Table 2: Impact of different replacement options using modules of higher efficiency.

If newer modules have higher efficiency, partial replacement with new modules can compensate for the lower output of remaining modules. This depends  on the ability to identify modules efficiencies, may be possible if string output monitoring is used and could be the basis.

As an example assume the existing modules have degraded to 80% of nominal capacity. Table 2 shows the impact of different replacement options using  modules of higher efficiency. The addition of storage to renewable energy systems will add another dimension to the decision.


[1] AKS Jardine: “Maintenance replacement and reliability”, 1994.

[2] DC Jordan and SR Kurtz: “Photovoltaic degradation rates — an analytical review”, National renewable energy laboratory, (NREL/JA-5200-51664), June 2012.

Send your comments to:

Related Articles

  • South African Government COVID-19 Corona Virus Resource Portal
  • Ministerial determinations propose 13813 MW of new-build by IPPs, none by Eskom
  • Crunch time for South Africa’s national nuclear company, Necsa
  • Dealing with the elephant in the room that is Eskom…
  • Interview with Minerals & Energy Minister Gwede Mantashe